Another expert opinion

National Post - In his testimony, Reilly acknowledged that the reason people died after receiving a Taser jolt remains unknown. "We're groping in an environment where we don't have all the facts," Reilly said. Still, in his experience, he said, there was a probability of "well less than 1%" of harm or death from Taser use. [LINK]

First of all, he has nothing to back up including harm in this accounting. So far as I am aware, we're still working on the questions of significant injury and 'death-by-taser'. The term harm is overly-broad and nearly meaningless. Harm might include long-term harm. The M26 and X26 tasers have been around less than a decade. Harm is a toss-away useless distraction. If you include damaging the public trust of the police, then the harm rate is nearly 100%.

On the assumption that he is lumping all tasers, M26 and X26, in together (mistake number 1), and he is probably including those "FAKE" taser-in-the-back demonstrations in the denominator (mistake number 2), and possibly even including the non-tasering taser incidents (denominator washing at its most blatant, I assume he isn't...), then perhaps his assessment of the risk of death (not including harm) of being under 1% might be roughly in the correct order of magnitude.

Remove the denominator washing, and "well less than 1%" might become roughly 1% adjusted to represent the risk from a full-on X26 tasering across the chest. THAT's the risk we're interested in; everything else is just luck-of-the-draw.

Even the most outspoken taser critic probably wouldn't want to try to defend that the taser-associated death rate being as high as 10%. Obviously that sort of number is simply not supported by the facts. But 0.1% for a properly weighted and compensated rate seems perfectly plausible (depending on how you look at the raw death rate and how much you trust the coroner's findings where Taser has had too much influence). The truth is (in my opinion) somewhere between those two values (in other words, within sight of 1% either way) for full-on X26 chest hits.

So Reilly is being more-or-less reasonable. He might not be perfectly accurate, and he may be washing out the denominator a bit as compared to full-on X26 hits across the chest, but at least he's not being silly.

On the other hand, Webster's 6 in a million, worst case is just being plain silly.